Migrating to MLflow v3.11.1
Version v3.11.1 introduces 4 breaking changes. This guide details how to update your code.
Released: 4/8/2026
⚠️ Check Your Code
If you use any of these symbols, you need to read this guide:
mlflow-tracing@mlflow/coremlflow-openai@mlflow/openaimlflow-anthropic@mlflow/anthropicmlflow-gemini@mlflow/geminiMLFLOW_ENABLE_INCREMENTAL_SPAN_EXPORTlitellmgepaRegistered Model namesMetaPromptOptimizerDatabricks API callscreate_experimentset_experimentlog_spans()mlflow.genai.evaluateuvtorch.exportskopsMLFLOW_ALLOW_PICKLE_DESERIALIZATIONBedrock providerConverse APIVertex AI gateway providerDatabricks gateway providerOllama gateway providerxAI (Grok) gateway providermlflow.tracing.context()autologgingpolars dataset supportDSPy trace chat viewBreaking Changes
●Issue #1
The MLflow TypeScript SDK packages have been renamed to use npm organization scoping. Update `package.json` dependencies and import statements: `mlflow-tracing` → `@mlflow/core`, `mlflow-openai` → `@mlflow/openai`, `mlflow-anthropic` → `@mlflow/anthropic`, `mlflow-gemini` → `@mlflow/gemini`. All packages are now at version `0.2.0`.
●Issue #2
The environment variable `MLFLOW_ENABLE_INCREMENTAL_SPAN_EXPORT` has been removed.
●Issue #3
The packages `litellm` and `gepa` have been removed from the `genai` extras.
●Issue #4
Registered Model names can no longer contain `/` or `:` characters.
Migration Steps
- 1If using the TypeScript SDK, update your `package.json` dependencies and import statements according to the new npm organization scoping: `mlflow-tracing` → `@mlflow/core`, `mlflow-openai` → `@mlflow/openai`, `mlflow-anthropic` → `@mlflow/anthropic`, `mlflow-gemini` → `@mlflow/gemini`.
- 2If you were relying on the `MLFLOW_ENABLE_INCREMENTAL_SPAN_EXPORT` environment variable, remove it.
- 3If you were using `litellm` or `gepa` via the `genai` extras, you may need to adjust your setup as they have been removed from those extras.
- 4Ensure Registered Model names do not contain `/` or `:` characters.
Release Summary
MLflow 3.11.1 introduces major features like AI-powered issue detection in traces, comprehensive AI Gateway budget controls, and a new Trace Graph View. This release also enhances security and reproducibility with pickle-free serialization and native UV dependency support.
Need More Details?
View the full release notes and all changes for MLflow v3.11.1.
View Full Changelog